Add link to paper and project page

#1
by nielsr HF Staff - opened
Files changed (1) hide show
  1. README.md +38 -20
README.md CHANGED
@@ -1,26 +1,26 @@
1
-
2
-
3
-
4
  ---
 
 
5
  library_name: transformers
6
  license: other
7
  license_name: nvidia-open-model-license
8
- license_link: >-
9
- https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/
10
  pipeline_tag: text-generation
11
- language:
12
- - en
13
  tags:
14
- - nvidia
15
- - reasoning
16
- - math
17
- - code
18
- - reinforcement learning
19
- - pytorch
20
  ---
21
 
22
  # AceReason-Nemotron: Advancing Math and Code Reasoning through Reinforcement Learning
23
 
 
 
 
 
24
  <p align="center">
25
 
26
  [![Technical Report](https://img.shields.io/badge/2505.16400-Technical_Report-blue)](https://arxiv.org/abs/2505.16400)
@@ -111,15 +111,33 @@ response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
111
  question = "" # code question
112
  starter_code = "" # starter code function header
113
 
114
- code_instruction_nostartercode = """Write Python code to solve the problem. Please place the solution code in the following format:\n```python\n# Your solution code here\n```"""
115
- code_instruction_hasstartercode = """Please place the solution code in the following format:\n```python\n# Your solution code here\n```"""
 
 
 
 
 
 
116
  if starter_code != "":
117
- question += "\n\n" + "Solve the problem starting with the provided function header.\n\nFunction header:\n" + "```\n" + starter_code + "\n```"
118
- question += "\n\n" + code_instruction_hasstartercode
 
 
 
 
 
 
 
 
 
119
  else:
120
- question += "\n\n" + code_instruction_nostartercode
 
 
121
 
122
- final_prompt = "<|User|>" + question + "<|Assistant|><think>\n"
 
123
  ```
124
  4. Our inference engine for evaluation is **vLLM==0.7.3** using top-p=0.95, temperature=0.6, max_tokens=32768.
125
 
@@ -143,4 +161,4 @@ Your use of this model is governed by the [NVIDIA Open Model License](https://ww
143
  journal={arXiv preprint arXiv:2505.16400},
144
  year={2025}
145
  }
146
- ```
 
 
 
 
1
  ---
2
+ language:
3
+ - en
4
  library_name: transformers
5
  license: other
6
  license_name: nvidia-open-model-license
7
+ license_link: https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/
 
8
  pipeline_tag: text-generation
 
 
9
  tags:
10
+ - nvidia
11
+ - reasoning
12
+ - math
13
+ - code
14
+ - reinforcement learning
15
+ - pytorch
16
  ---
17
 
18
  # AceReason-Nemotron: Advancing Math and Code Reasoning through Reinforcement Learning
19
 
20
+ This repository contains the AceReason-Nemotron-1.1 7B model presented in [AceReason-Nemotron 1.1: Advancing Math and Code Reasoning through SFT and RL Synergy](https://huggingface.co/papers/2506.13284).
21
+
22
+ Project page: https://huggingface.co/nvidia/AceReason-Nemotron-1.1-7B
23
+
24
  <p align="center">
25
 
26
  [![Technical Report](https://img.shields.io/badge/2505.16400-Technical_Report-blue)](https://arxiv.org/abs/2505.16400)
 
111
  question = "" # code question
112
  starter_code = "" # starter code function header
113
 
114
+ code_instruction_nostartercode = """Write Python code to solve the problem. Please place the solution code in the following format:
115
+ ```python
116
+ # Your solution code here
117
+ ```"""
118
+ code_instruction_hasstartercode = """Please place the solution code in the following format:
119
+ ```python
120
+ # Your solution code here
121
+ ```"""
122
  if starter_code != "":
123
+ question += "
124
+
125
+ " + "Solve the problem starting with the provided function header.
126
+
127
+ Function header:
128
+ " + "```
129
+ " + starter_code + "
130
+ ```"
131
+ question += "
132
+
133
+ " + code_instruction_hasstartercode
134
  else:
135
+ question += "
136
+
137
+ " + code_instruction_nostartercode
138
 
139
+ final_prompt = "<|User|>" + question + "<|Assistant|><think>
140
+ "
141
  ```
142
  4. Our inference engine for evaluation is **vLLM==0.7.3** using top-p=0.95, temperature=0.6, max_tokens=32768.
143
 
 
161
  journal={arXiv preprint arXiv:2505.16400},
162
  year={2025}
163
  }
164
+ ```